ParameterStore Class Reference

Interface for object which can store parameters. More...

Detailed Description

Interface for object which can store parameters.

#include <ParameterStore.h>

Inheritance diagram for ParameterStore:

Public Member Functions
	ParameterStore (OptionSet options, NEML2Object *object)
	ParameterStore (const ParameterStore &)=delete
	ParameterStore (ParameterStore &&)=delete
ParameterStore &	operator= (const ParameterStore &)=delete
ParameterStore &	operator= (ParameterStore &&)=delete
virtual	~ParameterStore ()=default
std::map< std::string, const VariableBase * >	_nl_params
	Map from nonlinear parameter names to their corresponding variable views.
std::map< std::string, Model * >	_nl_param_models
	Map from nonlinear parameter names to models which evaluate them.
const Storage< std::string, TensorValueBase > &	named_parameters () const
Storage< std::string, TensorValueBase > &	named_parameters ()
void	set_parameter (const std::string &, const Tensor &)
	}@
void	set_parameters (const std::map< std::string, Tensor > &)
	Set values for parameters.
TensorValueBase &	get_parameter (const std::string &name)
	Get a writable reference of a parameter.
const TensorValueBase &	get_parameter (const std::string &name) const
	Get a read-only reference of a parameter.
bool	has_nl_param () const
	Whether this parameter store has any nonlinear parameter.
const VariableBase *	nl_param (const std::string &) const
	Query the existence of a nonlinear parameter.
virtual std::map< std::string, const VariableBase * >	named_nonlinear_parameters (bool recursive=false) const
	Get all nonlinear parameters.
virtual std::map< std::string, Model * >	named_nonlinear_parameter_models (bool recursive=false) const
	Get all nonlinear parameters' models.
virtual void	send_parameters_to (const torch::TensorOptions &options)
	Send parameters to options.
template<typename T , typename = typename std::enable_if_t<std::is_base_of_v<TensorBase<T>, T>>>
const T &	declare_parameter (const std::string &name, const T &rawval)
	Declare a parameter.
template<typename T , typename = typename std::enable_if_t<std::is_base_of_v<TensorBase<T>, T>>>
const T &	declare_parameter (const std::string &name, const CrossRef< T > &crossref, bool allow_nonlinear)
	Declare a parameter.
template<typename T , typename = typename std::enable_if_t<std::is_base_of_v<TensorBase<T>, T>>>
const T &	declare_parameter (const std::string &name, const std::string &input_option_name, bool allow_nonlinear=false)
	Declare a parameter.

Constructor & Destructor Documentation

ParameterStore() [1/3]

ParameterStore	(	OptionSet	options,
		NEML2Object *	object )

ParameterStore() [2/3]

ParameterStore ( const ParameterStore & )

delete

ParameterStore() [3/3]

ParameterStore ( ParameterStore && )

delete

~ParameterStore()

virtual ~ParameterStore ( )

virtualdefault

Member Function Documentation

declare_parameter() [1/3]

template<typename T , typename >

const T & declare_parameter ( const std::string & name, const CrossRef< T > & crossref, bool allow_nonlinear )

protected

Declare a parameter.

Similar to the previous method, but additionally handles the resolution of cross-referenced parameters. Two attempts are made sequentially: first, the method tries to resolve CrossRef<T> into T directly; if that fails, the method tries to resolve CrossRef<T> into a nonlinear parameter where the raw string stored in the cross-ref is treated as the name of the model that defines the nonlinear parameter.

Template Parameters

T	Parameter type. See Statically shaped tensors for supported types.

Parameters

name	Name of the model parameter.
crossref	The cross-ref'ed "string" that defines the value of the model parameter.
allow_nonlinear	Whether allows coupling with a nonlinear parameter

Returns

T The value of the registered model parameter.

declare_parameter() [2/3]

template<typename T , typename >

const T & declare_parameter ( const std::string & name, const std::string & input_option_name, bool allow_nonlinear = false )

protected

Declare a parameter.

Similar to the previous methods, but this method takes care of the high-level logic to directly construct a (possibly nonlinear) parameter from the input option.

Template Parameters

T	Parameter type. See Statically shaped tensors for supported types.

Parameters

name	Name of the model parameter.
input_option_name	Name of the input option that defines the value of the model parameter.
allow_nonlinear	Whether allows coupling with a nonlinear parameter

Returns

T The value of the registered model parameter.

declare_parameter() [3/3]

template<typename T , typename >

const T & declare_parameter ( const std::string & name, const T & rawval )

protected

Declare a parameter.

Note that all parameters are stored in the host (the object exposed to users). An object may be used multiple times in the host, and the same parameter may be declared multiple times. That is allowed, but only the first call to declare_parameter constructs the parameter value, and subsequent calls only returns a reference to the existing parameter.

Template Parameters

T	Buffer type. See Statically shaped tensors for supported types.

Parameters

name	Buffer name
rawval	Buffer value

Returns

Reference to buffer

get_parameter() [1/2]

TensorValueBase & get_parameter

(

const std::string &

name

)

Get a writable reference of a parameter.

get_parameter() [2/2]

const TensorValueBase & get_parameter

(

const std::string &

name

)

const

Get a read-only reference of a parameter.

has_nl_param()

bool has_nl_param ( ) const

inline

Whether this parameter store has any nonlinear parameter.

named_nonlinear_parameter_models()

std::map< std::string, Model * > named_nonlinear_parameter_models ( bool recursive = false ) const

virtual

Get all nonlinear parameters' models.

Reimplemented in ComposedModel.

named_nonlinear_parameters()

std::map< std::string, const VariableBase * > named_nonlinear_parameters ( bool recursive = false ) const

virtual

Get all nonlinear parameters.

Reimplemented in ComposedModel.

named_parameters() [1/2]

Storage< std::string, TensorValueBase > & named_parameters

(

)

named_parameters() [2/2]

const Storage< std::string, TensorValueBase > & named_parameters ( ) const

inline

Returns

the buffer storage

nl_param()

const VariableBase * nl_param

(

const std::string &

name

)

const

Query the existence of a nonlinear parameter.

Returns

const VariableBase* Pointer to the VariableBase if the parameter associated with the given parameter name is nonlinear. Returns nullptr otherwise.

operator=() [1/2]

ParameterStore & operator= ( const ParameterStore & )

delete

operator=() [2/2]

ParameterStore & operator= ( ParameterStore && )

delete

send_parameters_to()

void send_parameters_to ( const torch::TensorOptions & options )

protectedvirtual

Send parameters to options.

Parameters

options

The target options

set_parameter()

void set_parameter	(	const std::string &	name,
		const Tensor &	value )

}@

Set the value for a parameter

set_parameters()

void set_parameters

(

const std::map< std::string, Tensor > &

param_values

)

Set values for parameters.

Member Data Documentation

_nl_param_models

std::map<std::string, Model *> _nl_param_models

protected

Map from nonlinear parameter names to models which evaluate them.

_nl_params

std::map<std::string, const VariableBase *> _nl_params

protected

Map from nonlinear parameter names to their corresponding variable views.